Search

A theoretical and experimental comparison of photon-counting cameras and CCD's for use in astronomical speckle imaging was performed. Photon-counting cameras able to detect single-photon events typically exhibit a lower quantum efficiency (QE) and suffer saturation effects at high light levels. In contrast, CCD's offer a high QE and virtually unlimited photon-count rate. However CCD's are limited at lower light levels by noise associated with the readout process. Speckle-imaging performance was quantified by derivation of the signal-to-noise ratio (SNR) of the power spectrum and the Knox-Thompson product to include CCD readout noise. Ground-based telescope observations at various light levels were obtained with an advanced, high-speed, low-noise CCD camera to verify SNR expressions. The useful operating ranges for these two camera types were compared by consideration of the effects of QE, readout noise, and maximum photon-count rate. Although photon-counting cameras continued to dominate low-light-level applications, CCD's are shown to offer significant improvements over photon-counting cameras for a wide range of light levels. Future reductions of readout noise will further improve CCD speckle-imaging performance. OCIS codes: 110.6150, 010.1330, 030.4280, 040.1520, 030.6610.

Optical wave-front propagation in a layered model for the atmosphere is analyzed by the use of diffraction theory, leading to a novel approach for utilizing artificial guide stars. Considering recent observations of layering in the atmospheric turbulence, the results of this paper indicate that, even for very large telescopes, a substantial enlargement of the compensated angular field of view is possible when two adaptive mirrors and four or five artificial guide stars are employed. The required number of guide stars increases as the thickness of the turbulent layers increases, converging to the conventional results at the limit of continuously turbulent atmosphere.

A new technique for generation of binary masks for the fabrication of diffractive optical elements is investigated. This technique, which uses commercially available desktop-publishing hardware and software in conjunction with a standard photoreduction camera, is much faster and less expensive than the conventional methods. The short turnaround time and low cost should give researchers a much greater degree of flexibility in the field of binary optics and enable wider application of diffractive-optics technology. Techniques for generating optical elements by using standard software packages that produce PostScript output are described. An evaluation of the dimensional fidelity of the mask reproduction from design to its realization in photoresist is presented.

Limits to the recovery of planetary images by speckle imaging were investigated by means of a numerical simulation of the image-forming process. Laboratory measurements established that the numerical model correctly represented the process. With this numerical model we studied the observing conditions required to obtain useful planetary data for Neptune and Io, and we learned which factors are important for the successful recovery of images. We also showed that our methods of amplitude and phase recovery do not contribute to image degradation, which is essentially caused by photon noise alone. From this conclusion we were led to two simple models of speckle imaging that involve neither details of the recovery procedure nor of the atmospheric behavior (except for the size of the seeing disk). Both models are governed by a dimensionless ratio that involves all parameters relevant to the process of image restoration. The models can be used by any observer, and they have been used by us to predict the performance of speckle imaging for T-tauri stars and for a Mars cloud study.